FastInstShadow: A Simple Query-Based Model for Instance Shadow Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Inoue, Takeru, Miyamoto, Ryusuke
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910867641073664
author Inoue, Takeru
Miyamoto, Ryusuke
author_facet Inoue, Takeru
Miyamoto, Ryusuke
contents Instance shadow detection is the task of detecting pairs of shadows and objects, where existing methods first detect shadows and objects independently, then associate them. This paper introduces FastInstShadow, a method that enhances detection accuracy through a query-based architecture featuring an association transformer decoder with two dual-path transformer decoders to assess relationships between shadows and objects during detection. Experimental results using the SOBA dataset showed that the proposed method outperforms all existing methods across all criteria. This method makes real-time processing feasible for moderate-resolution images with better accuracy than SSISv2, the most accurate existing method. Our code is available at https://github.com/wlotkr/FastInstShadow.
format Preprint
id arxiv_https___arxiv_org_abs_2503_07517
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FastInstShadow: A Simple Query-Based Model for Instance Shadow Detection
Inoue, Takeru
Miyamoto, Ryusuke
Computer Vision and Pattern Recognition
Instance shadow detection is the task of detecting pairs of shadows and objects, where existing methods first detect shadows and objects independently, then associate them. This paper introduces FastInstShadow, a method that enhances detection accuracy through a query-based architecture featuring an association transformer decoder with two dual-path transformer decoders to assess relationships between shadows and objects during detection. Experimental results using the SOBA dataset showed that the proposed method outperforms all existing methods across all criteria. This method makes real-time processing feasible for moderate-resolution images with better accuracy than SSISv2, the most accurate existing method. Our code is available at https://github.com/wlotkr/FastInstShadow.
title FastInstShadow: A Simple Query-Based Model for Instance Shadow Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.07517